The automatic generation of musical sounds from a musical score defined by a sequence of notes lacks variability. As a result, such automatically generated sound lacks the intonations that naturally result when different musicians add their own accent to the manner in which the notes are played.